A plurality of process data accumulates in the operation of paint shops for painting motor vehicle body components, for example the current position of application and painting robots, the speed of rotary atomisers, the charging voltage of electrostatic application devices or the flow rate of the individual media (e.g. paint and guide air), to name only some of the process data. Process data logging during continuous operation of the paint shop facilitates error analysis using the stored process data when a data error causes a malfunction in the paint shop.
Firstly, the Primas system, in which the process data accumulating during continuous operation of the paint shop are permanently and indefinitely stored in an external mass memory, installed by the operator of the paint shop, is used to log the process data. An initial disadvantage of said Primas system is the large memory requirement for storing the process data. A further disadvantage of the Primas system is the fact that the individual process data are logged and correspondingly stored without time synchronisation. For example, the control system of painting and application robots may have a slightly different system time from the conveyor control unit which conveys the motor vehicle body components to be painted through the paint shop. Such time synchronisation errors hinder later error analysis using the stored process data, as no precise time allocation of the process data is possible.
Secondly, in practice the process data accumulating are logged by the Softlog system or by data recorders. However, the process data to be logged must first be specified, to form a channel, before recording. When logging with data recorders, it is even necessary to lay cables before recording, the number of signals and the process data to be recorded being restricted. Such systems are usually only used for fault-finding for series errors and are not present in every paint shop. Furthermore, the duration of recording and the number of recording channels are restricted.
The Primas and Softlog systems both have the disadvantage that the process data accumulating is recorded independently of an error in the paint shop.
The Primas and Softlog systems also have the disadvantage that intermittent errors can only be analysed with extensive outlay, a lot of time being required for fault finding.
DE 32 24 586 A1 generally discloses an operating data logging method which can be used in the automotive industry. The operating data accumulating are continuously written to a memory. The accumulated, stored operating data can then be examined on a monitor in subsequent analysis of a recording period. A large memory is therefore required, as with the aforementioned, known Primas system. Furthermore, error analysis is also difficult in this case, as attribution of the operating data to an error is difficult.
A method of inspecting the quality of coating in a paint shop is disclosed in DE 100 36 741 A1, in which any errors may be saved in a database for each workpiece at the time of coating. Storage in the database is thus workpiece- and not error-related. Moreover, the database contains a plurality of errors which have occurred during the coating period, which hinders error analysis.
Finally, a system for determining the cause of errors, which is to be used, for example, in hot-rolling mills, is disclosed in WO 02/013015 A1. However, attribution of an error occurring to the appropriate process data sets is difficult in this case too.
Accordingly, there is a need for improving the above-mentioned error logging methods.